Face Generation¶

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data¶

You'll be using two datasets in this project:

MNIST
CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)

Found mnist Data
Found celeba Data

Explore the Data¶

MNIST¶

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')

<matplotlib.image.AxesImage at 0x7f0684bebb38>

CelebA¶

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))

<matplotlib.image.AxesImage at 0x7f06477f1cf8>

Preprocess the Data¶

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network¶

You'll build the components necessary to build a GANs by implementing the following functions below:

model_inputs
discriminator
generator
model_loss
model_opt
train

Check the Version of TensorFlow and Access to GPU¶

This will check to make sure you have the correct version of TensorFlow and access to a GPU

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))

TensorFlow Version: 1.0.0
Default GPU Device: /gpu:0

Input¶

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
Z input placeholder with rank 2 using z_dim.
Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    real_inputs = tf.placeholder(tf.float32, (None, image_width, image_height, image_channels),'real_inputs')
    input_z = tf.placeholder(tf.float32, (None, z_dim), 'input_z')
    learning_rate = tf.placeholder(tf.float32, name='learning_rate')

    return real_inputs, input_z, learning_rate

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)

Tests Passed

Discriminator¶

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

def discriminator(images, reuse=False, alpha=0.2):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    with tf.variable_scope('discriminator', reuse=reuse):
        # Input layer is 28x28x3
        x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        relu1 = tf.maximum(alpha * x1, x1)
        # 14x14x64
        
        x2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same')
        bn2 = tf.layers.batch_normalization(x2, training=True)
        relu2 = tf.maximum(alpha * bn2, bn2)
        # 7x7x128
        
        x3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same')
        bn3 = tf.layers.batch_normalization(x3, training=True)
        relu3 = tf.maximum(alpha * bn3, bn3)
        # 4x4x256

        # Flatten it
        flat = tf.reshape(relu3, (-1, 4*4*256))
        logits = tf.layers.dense(flat, 1)
        output = tf.sigmoid(logits)
        
        
        #print ("discriminator - images shape: {}".format (images.shape))
        #print ("discriminator - x1 shape: {}".format (x1.shape))
        #print ("discriminator - x2 shape: {}".format (x2.shape))
        #print ("discriminator - x3 shape: {}".format (x3.shape))
        #print ("discriminator - logits shape: {}".format (logits.shape)) 
        
        return output, logits   
    
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)

Tests Passed

Generator¶

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

def generator(z, out_channel_dim, is_train=True, alpha = 0.2):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
  
    # TODO: Implement Function
    reuse = not is_train

    with tf.variable_scope("generator", reuse = reuse):
        
        x1 = tf.layers.dense(z, 7*7*128)
        x1 = tf.reshape(x1, (-1, 7, 7, 128))
        bn1 = tf.layers.batch_normalization(x1, training=is_train)
        relu1 = tf.maximum(0.3*bn1, bn1)
        
#        # img size is 7*7*128
        x2 = tf.layers.conv2d_transpose(relu1, 64, 3, strides=2, padding="same")
        bn2 = tf.layers.batch_normalization(x2, training=is_train)
        relu2 = tf.maximum(0.3*bn2, bn2)
        
        # img size is 14*14*64
        x3 = tf.layers.conv2d_transpose(relu2, 32, 3, strides=1, padding="same")
        bn3 = tf.layers.batch_normalization(x3, training=is_train)
        relu3 = tf.maximum(0.3*bn3, bn3)
        
         # img size is 14*14*32
        x4 = tf.layers.conv2d_transpose(relu3, 16, 3, strides=2, padding="same")
        bn4 = tf.layers.batch_normalization(x4, training=is_train)
        relu4 = tf.maximum(0.3*bn4, bn4)
        
#        # img size is 28*28*16
        logits = tf.layers.conv2d_transpose(relu4, out_channel_dim, 3, strides=1, padding="same")
        # img size is 28*28*out_channel_dim
        
        output = tf.tanh(logits)
        #print ("generator - z shape: {}".format (z.shape))
        #print ("generator - x1 shape: {}".format (x1.shape))
        #print ("generator - x2 shape: {}".format (x2.shape))
        #print ("generator - x3 shape: {}".format (x3.shape))
        #print ("generator - logits shape: {}".format (logits.shape))    
    
        return output


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)

Tests Passed

Loss¶

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

discriminator(images, reuse=False)
generator(z, out_channel_dim, is_train=True)

def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # TODO: Implement Function
    
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)


    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))

    d_loss = d_loss_real + d_loss_fake

    return d_loss, g_loss
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)

Tests Passed

Optimization¶

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    # Optimize
    with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt
    



"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)

Tests Passed

Neural Network Training¶

Show Output¶

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train¶

Implement train to build and train the GANs. Use the following functions you implemented:

model_inputs(image_width, image_height, image_channels, z_dim)
model_loss(input_real, input_z, out_channel_dim)
model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode ):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # TODO: Build Model
    input_real, input_z, lr = model_inputs(data_shape[1], data_shape[2], data_shape[3], z_dim)
    d_loss, g_loss = model_loss(input_real, input_z, data_shape[3])
    d_train_opt, g_train_opt = model_opt(d_loss, g_loss, lr, beta1) 
    
    steps = 0
    print_every = 1
    show_every = 10
    losses = []
    n_images = 25
    
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                # TODO: Train Model
                steps += 1
                
                # Sample random noise for G
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))
                
                # Run optimizers
                _ = sess.run(d_train_opt, feed_dict={input_real: batch_images, input_z: batch_z, lr: learning_rate})
                _ = sess.run(g_train_opt, feed_dict={input_z: batch_z,input_real: batch_images,lr: learning_rate})
                
                
                if steps % print_every == 0:      
                    train_loss_d = d_loss.eval({input_z: batch_z, input_real: batch_images})
                    train_loss_g = g_loss.eval({input_z: batch_z})

                    print("Epoch {}/{}...".format(epoch_i+1, epoch_count),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                
                if steps % show_every == 1:
                    show_generator_output(sess,n_images, input_z, data_shape[3], data_image_mode)
 
                       
           # Save losses to view after training
           #losses.append((train_loss_d, train_loss_g))



    #return losses, samples

MNIST¶

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

batch_size = 128
z_dim = 100
learning_rate = 0.002
beta1 = 0.5


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)

Epoch 1/2... Discriminator Loss: 3.3540... Generator Loss: 7.5612

Epoch 1/2... Discriminator Loss: 16.7861... Generator Loss: 0.0002
Epoch 1/2... Discriminator Loss: 10.5772... Generator Loss: 0.0841
Epoch 1/2... Discriminator Loss: 6.6227... Generator Loss: 9.7045
Epoch 1/2... Discriminator Loss: 9.7297... Generator Loss: 0.0006
Epoch 1/2... Discriminator Loss: 2.9532... Generator Loss: 0.2367
Epoch 1/2... Discriminator Loss: 4.2240... Generator Loss: 4.3645
Epoch 1/2... Discriminator Loss: 3.2993... Generator Loss: 0.1564
Epoch 1/2... Discriminator Loss: 1.5167... Generator Loss: 0.9742
Epoch 1/2... Discriminator Loss: 1.5036... Generator Loss: 0.5421
Epoch 1/2... Discriminator Loss: 1.6496... Generator Loss: 0.7914

Epoch 1/2... Discriminator Loss: 1.7634... Generator Loss: 0.3488
Epoch 1/2... Discriminator Loss: 1.5144... Generator Loss: 3.1500
Epoch 1/2... Discriminator Loss: 3.1492... Generator Loss: 0.0878
Epoch 1/2... Discriminator Loss: 1.9621... Generator Loss: 1.3837
Epoch 1/2... Discriminator Loss: 2.6338... Generator Loss: 0.2301
Epoch 1/2... Discriminator Loss: 1.7695... Generator Loss: 2.2866
Epoch 1/2... Discriminator Loss: 2.0655... Generator Loss: 0.2193
Epoch 1/2... Discriminator Loss: 0.8610... Generator Loss: 2.5050
Epoch 1/2... Discriminator Loss: 3.3058... Generator Loss: 0.1627
Epoch 1/2... Discriminator Loss: 3.0922... Generator Loss: 2.8460

Epoch 1/2... Discriminator Loss: 2.5592... Generator Loss: 0.1293
Epoch 1/2... Discriminator Loss: 0.9939... Generator Loss: 6.2857
Epoch 1/2... Discriminator Loss: 0.3851... Generator Loss: 2.5291
Epoch 1/2... Discriminator Loss: 3.0611... Generator Loss: 0.1776
Epoch 1/2... Discriminator Loss: 1.5040... Generator Loss: 1.5890
Epoch 1/2... Discriminator Loss: 1.0971... Generator Loss: 0.7978
Epoch 1/2... Discriminator Loss: 0.8763... Generator Loss: 0.9781
Epoch 1/2... Discriminator Loss: 0.7714... Generator Loss: 1.2483
Epoch 1/2... Discriminator Loss: 0.7964... Generator Loss: 0.9932
Epoch 1/2... Discriminator Loss: 0.9096... Generator Loss: 0.9910

Epoch 1/2... Discriminator Loss: 1.2667... Generator Loss: 0.6393
Epoch 1/2... Discriminator Loss: 1.7716... Generator Loss: 1.3567
Epoch 1/2... Discriminator Loss: 2.7067... Generator Loss: 0.0862
Epoch 1/2... Discriminator Loss: 2.7629... Generator Loss: 7.2255
Epoch 1/2... Discriminator Loss: 2.1468... Generator Loss: 0.2725
Epoch 1/2... Discriminator Loss: 1.2072... Generator Loss: 1.0805
Epoch 1/2... Discriminator Loss: 1.5013... Generator Loss: 0.5880
Epoch 1/2... Discriminator Loss: 1.3448... Generator Loss: 0.8583
Epoch 1/2... Discriminator Loss: 1.5944... Generator Loss: 0.8554
Epoch 1/2... Discriminator Loss: 2.2720... Generator Loss: 0.1886

Epoch 1/2... Discriminator Loss: 5.3319... Generator Loss: 5.3482
Epoch 1/2... Discriminator Loss: 2.1692... Generator Loss: 0.1739
Epoch 1/2... Discriminator Loss: 0.9237... Generator Loss: 1.7571
Epoch 1/2... Discriminator Loss: 0.9174... Generator Loss: 1.2048
Epoch 1/2... Discriminator Loss: 1.1424... Generator Loss: 0.5163
Epoch 1/2... Discriminator Loss: 1.2816... Generator Loss: 1.6007
Epoch 1/2... Discriminator Loss: 1.7392... Generator Loss: 0.2634
Epoch 1/2... Discriminator Loss: 1.8240... Generator Loss: 1.7444
Epoch 1/2... Discriminator Loss: 1.9041... Generator Loss: 0.2416
Epoch 1/2... Discriminator Loss: 1.4360... Generator Loss: 2.0542

Epoch 1/2... Discriminator Loss: 1.3815... Generator Loss: 0.4345
Epoch 1/2... Discriminator Loss: 0.9801... Generator Loss: 1.6921
Epoch 1/2... Discriminator Loss: 1.2718... Generator Loss: 0.4750
Epoch 1/2... Discriminator Loss: 1.3175... Generator Loss: 1.9794
Epoch 1/2... Discriminator Loss: 1.8767... Generator Loss: 0.2111
Epoch 1/2... Discriminator Loss: 1.4911... Generator Loss: 3.3110
Epoch 1/2... Discriminator Loss: 0.9019... Generator Loss: 0.6949
Epoch 1/2... Discriminator Loss: 0.5165... Generator Loss: 1.6059
Epoch 1/2... Discriminator Loss: 0.6178... Generator Loss: 1.3517
Epoch 1/2... Discriminator Loss: 0.7756... Generator Loss: 0.9062

Epoch 1/2... Discriminator Loss: 1.1922... Generator Loss: 3.5827
Epoch 1/2... Discriminator Loss: 4.2425... Generator Loss: 0.0165
Epoch 1/2... Discriminator Loss: 3.2051... Generator Loss: 5.9150
Epoch 1/2... Discriminator Loss: 0.9287... Generator Loss: 0.7266
Epoch 1/2... Discriminator Loss: 0.8608... Generator Loss: 0.8386
Epoch 1/2... Discriminator Loss: 0.9886... Generator Loss: 2.0390
Epoch 1/2... Discriminator Loss: 0.9060... Generator Loss: 0.6875
Epoch 1/2... Discriminator Loss: 0.7443... Generator Loss: 1.7888
Epoch 1/2... Discriminator Loss: 0.7289... Generator Loss: 0.8629
Epoch 1/2... Discriminator Loss: 0.4942... Generator Loss: 2.6786

Epoch 1/2... Discriminator Loss: 0.4498... Generator Loss: 1.2155
Epoch 1/2... Discriminator Loss: 0.2526... Generator Loss: 2.2147
Epoch 1/2... Discriminator Loss: 0.2112... Generator Loss: 2.2781
Epoch 1/2... Discriminator Loss: 0.1680... Generator Loss: 2.3575
Epoch 1/2... Discriminator Loss: 0.1621... Generator Loss: 2.3842
Epoch 1/2... Discriminator Loss: 0.2453... Generator Loss: 1.9694
Epoch 1/2... Discriminator Loss: 0.3067... Generator Loss: 2.2017
Epoch 1/2... Discriminator Loss: 0.2873... Generator Loss: 1.7969
Epoch 1/2... Discriminator Loss: 0.7523... Generator Loss: 9.0632
Epoch 1/2... Discriminator Loss: 6.0938... Generator Loss: 0.0036

Epoch 1/2... Discriminator Loss: 2.5514... Generator Loss: 3.7484
Epoch 1/2... Discriminator Loss: 6.3155... Generator Loss: 0.0045
Epoch 1/2... Discriminator Loss: 2.7751... Generator Loss: 4.7347
Epoch 1/2... Discriminator Loss: 2.0296... Generator Loss: 0.9548
Epoch 1/2... Discriminator Loss: 1.5984... Generator Loss: 0.4429
Epoch 1/2... Discriminator Loss: 0.9994... Generator Loss: 1.5726
Epoch 1/2... Discriminator Loss: 0.7929... Generator Loss: 1.7956
Epoch 1/2... Discriminator Loss: 0.7506... Generator Loss: 1.0124
Epoch 1/2... Discriminator Loss: 0.6424... Generator Loss: 1.1920
Epoch 1/2... Discriminator Loss: 0.9155... Generator Loss: 1.1304

CelebA¶

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

batch_size = 128
z_dim = 100
learning_rate = 0.002
beta1 = 0.5


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)

Epoch 1/1... Discriminator Loss: 15.4814... Generator Loss: 0.0005

Epoch 1/1... Discriminator Loss: 0.9265... Generator Loss: 4.2445
Epoch 1/1... Discriminator Loss: 1.7310... Generator Loss: 2.4108
Epoch 1/1... Discriminator Loss: 0.8897... Generator Loss: 2.7849
Epoch 1/1... Discriminator Loss: 0.5604... Generator Loss: 14.8210
Epoch 1/1... Discriminator Loss: 6.2275... Generator Loss: 0.1163
Epoch 1/1... Discriminator Loss: 3.0866... Generator Loss: 23.6666
Epoch 1/1... Discriminator Loss: 0.2896... Generator Loss: 12.2903
Epoch 1/1... Discriminator Loss: 3.9086... Generator Loss: 1.2970
Epoch 1/1... Discriminator Loss: 3.2235... Generator Loss: 0.9078
Epoch 1/1... Discriminator Loss: 0.3666... Generator Loss: 8.1495

Epoch 1/1... Discriminator Loss: 0.3435... Generator Loss: 5.8124
Epoch 1/1... Discriminator Loss: 4.8763... Generator Loss: 0.2026
Epoch 1/1... Discriminator Loss: 7.2064... Generator Loss: 11.8029
Epoch 1/1... Discriminator Loss: 1.9231... Generator Loss: 2.0278
Epoch 1/1... Discriminator Loss: 2.5075... Generator Loss: 0.1652
Epoch 1/1... Discriminator Loss: 1.1988... Generator Loss: 2.2337
Epoch 1/1... Discriminator Loss: 1.3438... Generator Loss: 0.8345
Epoch 1/1... Discriminator Loss: 1.2253... Generator Loss: 1.0910
Epoch 1/1... Discriminator Loss: 1.6924... Generator Loss: 0.6484
Epoch 1/1... Discriminator Loss: 1.5104... Generator Loss: 4.3433

Epoch 1/1... Discriminator Loss: 3.5615... Generator Loss: 0.0924
Epoch 1/1... Discriminator Loss: 3.0920... Generator Loss: 7.1493
Epoch 1/1... Discriminator Loss: 0.9674... Generator Loss: 1.1972
Epoch 1/1... Discriminator Loss: 1.9040... Generator Loss: 0.3285
Epoch 1/1... Discriminator Loss: 1.3969... Generator Loss: 3.9094
Epoch 1/1... Discriminator Loss: 0.7535... Generator Loss: 1.2209
Epoch 1/1... Discriminator Loss: 1.3660... Generator Loss: 0.5486
Epoch 1/1... Discriminator Loss: 0.8750... Generator Loss: 2.6800
Epoch 1/1... Discriminator Loss: 1.3238... Generator Loss: 0.7300
Epoch 1/1... Discriminator Loss: 0.7957... Generator Loss: 2.7253

Epoch 1/1... Discriminator Loss: 1.0331... Generator Loss: 0.7773
Epoch 1/1... Discriminator Loss: 1.0546... Generator Loss: 4.0208
Epoch 1/1... Discriminator Loss: 1.5447... Generator Loss: 0.5389
Epoch 1/1... Discriminator Loss: 0.9445... Generator Loss: 6.2032
Epoch 1/1... Discriminator Loss: 1.3796... Generator Loss: 0.8598
Epoch 1/1... Discriminator Loss: 0.3154... Generator Loss: 4.9297
Epoch 1/1... Discriminator Loss: 0.2060... Generator Loss: 3.5531
Epoch 1/1... Discriminator Loss: 0.5691... Generator Loss: 1.3169
Epoch 1/1... Discriminator Loss: 0.2921... Generator Loss: 5.7580
Epoch 1/1... Discriminator Loss: 1.7129... Generator Loss: 0.3934

Epoch 1/1... Discriminator Loss: 7.2876... Generator Loss: 13.2007
Epoch 1/1... Discriminator Loss: 0.2341... Generator Loss: 3.0293
Epoch 1/1... Discriminator Loss: 1.0026... Generator Loss: 0.8249
Epoch 1/1... Discriminator Loss: 0.3610... Generator Loss: 1.9336
Epoch 1/1... Discriminator Loss: 0.2557... Generator Loss: 2.8648
Epoch 1/1... Discriminator Loss: 0.1768... Generator Loss: 2.8073
Epoch 1/1... Discriminator Loss: 0.1794... Generator Loss: 2.4720
Epoch 1/1... Discriminator Loss: 0.2153... Generator Loss: 2.2405
Epoch 1/1... Discriminator Loss: 0.1832... Generator Loss: 2.7765
Epoch 1/1... Discriminator Loss: 0.2370... Generator Loss: 2.2506

Epoch 1/1... Discriminator Loss: 0.2398... Generator Loss: 2.1944
Epoch 1/1... Discriminator Loss: 0.2413... Generator Loss: 2.5624
Epoch 1/1... Discriminator Loss: 0.4089... Generator Loss: 1.7415
Epoch 1/1... Discriminator Loss: 0.1763... Generator Loss: 4.4690
Epoch 1/1... Discriminator Loss: 0.0559... Generator Loss: 3.8846
Epoch 1/1... Discriminator Loss: 0.0927... Generator Loss: 2.7834
Epoch 1/1... Discriminator Loss: 0.2207... Generator Loss: 1.9226
Epoch 1/1... Discriminator Loss: 0.1316... Generator Loss: 2.4468
Epoch 1/1... Discriminator Loss: 0.1001... Generator Loss: 2.8413
Epoch 1/1... Discriminator Loss: 0.1269... Generator Loss: 2.5659

Epoch 1/1... Discriminator Loss: 0.1091... Generator Loss: 2.7435
Epoch 1/1... Discriminator Loss: 0.1397... Generator Loss: 2.4280
Epoch 1/1... Discriminator Loss: 0.1430... Generator Loss: 2.5285
Epoch 1/1... Discriminator Loss: 0.1621... Generator Loss: 2.4849
Epoch 1/1... Discriminator Loss: 0.1806... Generator Loss: 2.2746
Epoch 1/1... Discriminator Loss: 0.1731... Generator Loss: 2.4344
Epoch 1/1... Discriminator Loss: 0.1639... Generator Loss: 2.2467
Epoch 1/1... Discriminator Loss: 0.1872... Generator Loss: 2.4418
Epoch 1/1... Discriminator Loss: 0.5659... Generator Loss: 0.9909
Epoch 1/1... Discriminator Loss: 6.0628... Generator Loss: 7.6878

Epoch 1/1... Discriminator Loss: 6.2614... Generator Loss: 0.0042
Epoch 1/1... Discriminator Loss: 2.2868... Generator Loss: 4.5013
Epoch 1/1... Discriminator Loss: 1.7568... Generator Loss: 2.7455
Epoch 1/1... Discriminator Loss: 1.8423... Generator Loss: 0.7794
Epoch 1/1... Discriminator Loss: 1.7416... Generator Loss: 0.5075
Epoch 1/1... Discriminator Loss: 1.3864... Generator Loss: 2.1878
Epoch 1/1... Discriminator Loss: 1.0450... Generator Loss: 1.3008
Epoch 1/1... Discriminator Loss: 0.7939... Generator Loss: 1.0902
Epoch 1/1... Discriminator Loss: 0.9298... Generator Loss: 0.9996
Epoch 1/1... Discriminator Loss: 1.1418... Generator Loss: 0.9400

Epoch 1/1... Discriminator Loss: 1.0968... Generator Loss: 1.1294
Epoch 1/1... Discriminator Loss: 1.2178... Generator Loss: 0.9759
Epoch 1/1... Discriminator Loss: 1.6132... Generator Loss: 2.2988
Epoch 1/1... Discriminator Loss: 1.0895... Generator Loss: 0.8255
Epoch 1/1... Discriminator Loss: 0.7993... Generator Loss: 1.4857
Epoch 1/1... Discriminator Loss: 0.8061... Generator Loss: 1.2035
Epoch 1/1... Discriminator Loss: 0.7601... Generator Loss: 1.3171
Epoch 1/1... Discriminator Loss: 0.9484... Generator Loss: 0.9501
Epoch 1/1... Discriminator Loss: 1.1936... Generator Loss: 1.6412
Epoch 1/1... Discriminator Loss: 1.4540... Generator Loss: 0.8693

Submitting This Project¶

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.